A Facial Recognition Project Report
Item Preview
Share or Embed This Item
Flag this item for
- Publication date
- 1963-01
- Usage
- Public Domain Mark 1.0
- Topics
- computer vision, cv, cia, facial recognition
- Collection
- opensource; community
- Language
- English
These two studies constitute the first known attempt to implement a system to perform computerized facial recognition.
Both are authored by Dr. Woodrow Wilson Bledsoe for Panaromic Research, Inc, and addressed for submission to The King-Hurley Research Group. Importantly, the King-Hurley Research group was a front organization for the United States Central Intelligence Agency (CIA) from 1955 to 1969, according to Bill Richards, staff writer for the Washington Post in his May 12th 1978 article, "Firm's Suits Against CIA Shed Light on Clandestine Air Force". That the King-Hurley Research Group was a CIA front is also affirmed by Christopher Robbins in his 1979 book, "The Invisible Air Force" (1979 Mcmillion London Ltd, page 65).
The first report, ‘A PROPOSAL FOR A STUDY TO DETERMINE THE FEASIBILITY OF A SIMPLIFIED FACE RECOGNITION MACHINE’, dated 30 January 1963, aims to evaluate whether it is possible using existing technology to create a machine which can reliably solve a simplified face recognition problem: using only pictures from a single view (e.g. same angle & perspective) as well as high resolution pictures, identify the identity or a person from a picture that is not known to the machine.
The second report, ‘FACIAL RECOGNITION PROJECT REPORT’, includes the first report in its March 6th 1964 submission, and covers the expanded ambitions of the project to the more complex facial recognition problem: recognize a person's identity from novel photographs taken from multiple angles and differing lighting conditions. The report concludes that “picture recognition by machines for a large sample of people is beyond the state of the art of the present pattern recognition and computer technology at this time”.
The four 'lines of attack' carried out by the study are:
- developing more powerful recognition techniques for identifying the whole face at once
- dividing up the face into a sequence of parts or features, recognizing (or classifying) each of the figures, and combining the parts somehow for a recognition of the whole
- partitioning the categories in such a way that at any one time the machine is required to make decisions between only two (or a few) super categories. For example, the machine might first decide whether the person is male or female, then decide whether the person is Oriental or not, etc
- Preprocessing of the pictures to eliminate non-essential variability before the recognition process began.
The report includes a number of charts, graphs and photographic plates which, while quite degraded by photocopying, have legible and useful notation. This is particularly explanatory with regards to Bledsoe's discussion of rotating and transforming the angle of a picture to take into account a photographic angle that is different than the reference image.
Copies of the reports were tracked to an off-site storage facility for the Dolph Briscoe Center for American History of the University of Texas at Austin in 2012 and 2013.
Also of note: the CIA rejected a FOIA request issued to them in 2014, stating it could neither confirm nor deny their existence or nonexistence, citing various national security exceptions.
- Addeddate
- 2016-08-14 00:02:51
- Coverleaf
- 0
- Identifier
- firstfacialrecognitionresearch
- Identifier-ark
- ark:/13960/t5x689z6t
- Ocr
- ABBYY FineReader 11.0
- Ppi
- 200
- Scanner
- Internet Archive HTML5 Uploader 1.6.3
- Year
- 1963
comment
Reviews
There are no reviews yet. Be the first one to
write a review.
685 Views
DOWNLOAD OPTIONS
IN COLLECTIONS
Uploaded by annyuresearcher on